Understanding mechanistic causes of population change is critical for managing and conserving species. Integrated population models (IPMs) allow for quantifying population changes while directly relating environmental drivers to vital rates, but power of IPMs to detect trends and environmental effects on vital rates remains understudied. We simulated data for an IPM fewer than 41 scenarios to determine the power to detect trends and environmental effects on vital rates based on study duration, sample size, detection probability, and effect size. Our results indicated that temporal duration of a study and effect size, rather than sample size of each individual data set or detection probability, had the greatest influence on the power to identify trends in adult survival and fecundity. When using only 10 years of data, we were unable to identify a 50% increase in adult survival but were able to identify this increase with 22 years of data. When using only capture-recapture data in a traditional Cormack-Jolly-Seber analysis, we lacked sufficient power to identify trends in survival, and power of the Cormack-Jolly-Seber model was always less than the IPM. The IPM had greater power to identify trends and environmental effects on fecundity (e.g., we detected a 58% change in fecundity using 12 years of data). Models with effects of environmental variables on vital rates had less power than trends, likely to be due to increased annual variation in the vital rate when modeling responses to environmental effects that varied by year. Lack of power in the Cormack-Jolly-Seber analysis could be due to the relatively small variability in adult survival compared with fecundity, given the life history of our simulated species. As interannual variation in environmental conditions will probably increase with climate change, this type of analysis can help to inform the study duration needed, which may be a shifting target given future climate uncertainty and the complex nature of environmental correlations with demography.

中文翻译：

---

样本量、检测概率和研究持续时间对综合人口模型估计的相对影响。

了解人口变化的机制原因对于管理和保护物种至关重要。综合人口模型 (IPM) 允许量化人口变化，同时直接将环境驱动因素与生命率相关联，但 IPM 检测趋势和环境对生命率影响的能力仍未得到充分研究。我们模拟了 IPM 少于 41 种场景的数据，以确定根据研究持续时间、样本量、检测概率和效应量检测趋势和环境对生命率影响的能力。我们的结果表明，研究的时间持续时间和效应量，而不是每个单独数据集的样本量或检测概率，对识别成人存活率和繁殖力趋势的能力影响最大。当仅使用 10 年的数据时，我们无法确定成人存活率增加了 50%，但能够通过 22 年的数据确定这种增加。在传统的 Cormack-Jolly-Seber 分析中仅使用捕获-再捕获数据时，我们缺乏足够的能力来识别生存趋势，而 Cormack-Jolly-Seber 模型的能力始终低于 IPM。IPM 具有更大的能力来识别趋势和环境对繁殖力的影响（例如，我们使用 12 年的数据检测到繁殖力发生了 58% 的变化）。具有环境变量对生命率影响的模型的功效低于趋势，这可能是由于在对每年变化的环境影响的响应进行建模时，生命率的年度变化增加。考虑到我们模拟物种的生活史，Cormack-Jolly-Seber 分析中缺乏效力可能是由于与繁殖力相比，成年存活率的变异性相对较小。由于环境条件的年际变化可能会随着气候变化而增加，这种类型的分析有助于告知所需的研究持续时间，考虑到未来气候的不确定性和环境与人口统计学相关性的复杂性，这可能是一个不断变化的目标。